r"""Fix the valuehead checkpoint files. The model is already unwrapped. There are three cases: 1. full tuning without ds_zero3: state_di

r"""Convert dataset examples to messages.

r"""Get the base elements that are commonly used.

获取缓存文件路径

获取中国国家统计局的明细数据 参数： id: 统计项ID data_type: 月度/季度 wdcode: 固定值，zb cat: 分类数据

r"""Get a dataset converter.

r"""Freeze vision tower and language model for VLM full/freeze tuning.

Generate error message for token limit exceeded

r"""Get a list of scores of the reward model.

获取存储统计信息

Create a huggingface processor to process multimodal data. Args: name_or_path (str): The name of the processor. Returns: tra

Initialize connections to configured MCP servers.

r"""Load model config.

加载模型和分词器。

Mark agent as terminated

处理用户输入，动态解析是否需要调用MCP工具

处理用户查询，使用 OpenAI 客户端发送请求

r"""Remove thought from assistant message.

保存模型。 参数: save_path: 保存路径

获取股票介绍信息并保存到txt文件。 :param symbol: 股票代码 :param market: 市场类型 :param save_path: 保存的txt文件路径

基于关键字列表, 在项目库中搜索相关的内容

Set input tensor to be used instead of forward()'s input. When doing pipeline parallelism the input from the previous stage comes fro

Parses each jsonl line and yields it as a dictionary

Union two tensordicts.

TODO(zhangchi.usc1992): We may change the implementation later

r"""Add or replace eos token to the tokenizer.

Reorder the data on single controller such that each dp rank gets similar total tokens

Broadcast message to all active agents.

broadcast tensor in tp shards across mp_group

broadcast tensor in tp shards across mp_group

r"""Build a dictionary containing the current training configuration.

Build parameters schema based on response type.

Calculate tokens for high detail images based on dimensions

r"""Check if the video is nested images.

清理和标准化答案。 参数: answer: 原始答案 返回: 清理后的答案

r"""Remove args with NoneType or False or empty string value.

r"""Clean up the server process when the engine is destroyed.

r"""Convert elements to token ids.